Features of the built environment that support active living among seniors remain relatively unknown. Research in the transportation and urban planning fields indicates that people walk more in accessible neighborhoods. Specifically, accessible neighborhoods defined by more sidewalk continuity, easier street crossings, more public transit, accessible retail and other services, and availability of parks and walking trails may encourage walking for exercise or as a way to carry out routine daily activities. Ultimately, increased walking and overall physical activity should result in maintenance of physical function, especially lower extremity function. Existing research on neighborhood factors and physical activity is limited in several ways: (1) investigators have not identified specific features of neighborhood to measure a-priori based on a statement of the direct and indirect causal pathways that may be operating, (2) built environment measures are inconsistently operationalized, rely on self-report, and only assess ecologic neighborhood characteristics, and (3) cross-sectional measurement of physical activity and physical function that relies primarily on self-report. The proposed research will address these limitations and expand on existing knowledge by linking objectively measured physical activity in older adults with disaggregated and objective information on neighborhood accessibility. We propose to test hypotheses about the relation between neighborhood accessibility, walking behavior and physical activity, and lower-extremity function in a large cohort of older men. We will merge neighborhood characteristic data with physical activity and physical function data from 500 participants in the ongoing Osteoporotic Fractures in Men (MrOS) study living in the Portland OR metropolitan area. Our governmental transportation planning partner, Metro, will use a regional planning and transportation Geographic Information Systems (GIS) database in order to compute measures of neighborhood accessibility within a quarter-mile radius of the address of each study participant. Longitudinal physical activity and physical function data are measured using self-report and objective assessments. Linear regression and mixed effects (nested) models will be used to address the study hypotheses. The results of this research are an important first step in quantifying the relation between neighborhood accessibility and physical activity, especially walking, among the elderly.